Arthur Caplan’s lecture this morning was way better than I expected it would be–in addition to having what turned out to be a very intellectually stimulating topic, he was also really funny.
He talked about peer review and conflict of interest in an interesting way. Toward the beginning of his talk, he criticized peer review journals as being under the influence of big pharma–essentially that they are “on the take.” How, he wondered could librarians ethically recommend on-the-take peer review journals over Wikipedia, which is merely just the work of lunatics? (In a funnier way that I put it.)
He gave a couple of examples of recent journal articles that were essentially debased in the popular media for having conflict of interest problems. One of the examples, dear to my own heart, was the excellent JAMA fetal pain systematic review that was attacked by the religious right as being unethical because some of the authors were pro-choice and some received funding from Planned Parenthood and similar organizations. Apparently, Marcia Angell and another NEJM editor both publicly critiqued JAMA for not disclosing that history with the article, because if it was the other way around (an anti-choice person who found that fetal pain does start at 20 weeks), readers would want to know.
Caplan basically went on to say that this was an insane path to go down, because if everyone had to disclose every potential thing about themselves (their rivals in the field, high school enemies, political leanings, religion, etc), then journals would be all biography and no articles.
He said that there was a crisis of confidence in the scientific method–that the public couldn’t or didn’t trust science anymore because of the relativism inherent in this conflict of interest type of criticism. (It reminded me a bit of the D-LIB tagging article that talked about tagging being bad because it denied the Aristotelean unities.)
He suggested the following as solutions:
- open peer review (if not the reviewers, then at least to publish the reviews with the articles
- pay peer reviewers
- reward peer reviewers in tenure/pay scale systems
- peer reviewers should disclose conflicts of interest as well as authors
- hire statisticians to review research methodologies
- tone down media releases
- stop talking about medical breakthroughs–until something is repeated and proven it is newsworthy, but not a breakthrough
- make editorial process transparent to the public
Someone asked if he thought PLoS One was the way peer review should be headed, and he said yes–Web 2.0 technologies can have a big impact on improving transparency of the peer review process.